Another panel series/parallel question

Chesuncok1
Chesuncok1 Registered Users Posts: 5
Hello,
I am a newbie to this forum and have been reading stuff for days. Thank you so much for all the great info.

I have a 24v system, 8 T105's in a series/parallel configuration, 2800 watt Magnum inverter at a remote camp off grid.

I have 6 solar panels now and will be adding 6 more soon as the are on the way now. The 6 panels are configured (2 panels in series, 3 strings paralleled) as follows

2 Kyocera 120 watt in series with a Voc 21.5v, Vmp 16.9v, Isc 7.45a, Imp 7.1a these are paralleled with 2 strings of

2 DM solar 145 watt in series with a Voc 21.7v?, Vmp 18.0v, Isc 8.48a?, Imp 8.0a (I am sure of the Vmp & Imp, guessing on the Voc & Isc)

Based on what I have read here, connecting panels in series you look a the difference in current and when connecting in parallel, you look at voltage. The above configuration seemed to yield the most out of the panels. If I add 6 more DM solar 145 watt panels and configure the 6 additional panels as 3 more strings like above, I will have a current of 47 - 48 amps. If I go this way I will have to change my 70 foot run of #6 copper to something bigger which I am OK with if it makes sense. If I do could I use a #2 URD cable as it has 4 conductors (3 - #2's & 1 - #4) so I could do a parallel thing. I am planning on buying the MidNite Classic 150 but could buy the 200 or 250 if it makes sense to you guys that I instead put more panels in series. I didn't know how to configure because if I put 5 DM solar 145's and 1 Kyocera 120 in a string then wouldn't all panels in the string be capped at 7.1 amps which I would lose about 80 watts per string?

I greatly appreciate your thoughts
Scott

Comments

  • ggunn
    ggunn Solar Expert Posts: 1,973 ✭✭✭
    Re: Another panel series/parallel question
    Chesuncok1 wrote: »
    ...if I put 5 DM solar 145's and 1 Kyocera 120 in a string then wouldn't all panels in the string be capped at 7.1 amps which I would lose about 80 watts per string?

    Yes.

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Another panel series/parallel question

    If you go with the higher voltage Midnite controllers--You can put 6x Kyocera in series and 3x DM solar in series, then connect the two in parallel as the Vmp for the panels are very close together and will not measurably hurt your performance (for 1-2 strings, you don't usually need a combiner box. For three or more parallel strings, usually you do need the series protection "fuse" per string to protect against short circuits in any of the solar panels strings.).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Another panel series/parallel question

    How about four strings with three panels in each string? The advantage of this over six panels in series is that your charge controller will run cooler and more efficiently. It is easier for your controller to output 24 volts from a 54 volt input than from a 108 volt input.

    If you do four strings of three panels per string, your voltage drop over the 70 ft run will be about 3.3% (at full power, lesser drop at lower power). If you try to minimize the voltage drop by going six panels in series, you will lose whatever you gain by the decreased efficiency in your controller.

    The best way to minimize your losses would be to use a heavier cable for the 70 ft run. BTW, your losses with four strings are not extreme. You could get by with your existing cable.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Chesuncok1
    Chesuncok1 Registered Users Posts: 5
    Re: Another panel series/parallel question
    vtmaps wrote: »
    How about four strings with three panels in each string? The advantage of this over six panels in series is that your charge controller will run cooler and more efficiently. It is easier for your controller to output 24 volts from a 54 volt input than from a 108 volt input.

    If you do four strings of three panels per string, your voltage drop over the 70 ft run will be about 3.3% (at full power, lesser drop at lower power). If you try to minimize the voltage drop by going six panels in series, you will lose whatever you gain by the decreased efficiency in your controller.

    The best way to minimize your losses would be to use a heavier cable for the 70 ft run. BTW, your losses with four strings are not extreme. You could get by with your existing cable.

    --vtMaps

    Thank you for all the input.

    How can you figure the losses of the MidNite Classic 150? I like the idea of 3 panels in a string for 52 volts, how much more would I lose if I went 4 panels in a string for 72 volts? That would keep my loss in the wire run at 2.07 using the Southwire voltage drop calculator. The reason I ask this is because next year I need to move the batteries, controller, etc. 30' further away for a total of a 100'. There will be a junction box at the old location and I could use #4 or #2 if need be for the extra 30'. The only thing I found for the Classic 150 was the graph in their manual that look like it would still put out 94 amps using 72 volt PV charging a 24 volt battery bank but obviously the losses are going to be greater the further you separate the battery and PV voltage. I am wondering if there is a way to tell if the addition loss by going to 72 volts over 54 volts would be 10 watts, 50 watts, etc. It would be nice to only have to redo my wiring of the panels this one last time so I am all set for the 30' move next year.

    Thanks again for your help
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Another panel series/parallel question
    Chesuncok1 wrote: »
    How can you figure the losses of the MidNite Classic 150? I like the idea of 3 panels in a string for 52 volts, how much more would I lose if I went 4 panels in a string for 72 volts?

    I don't have the efficiency curves for the classic, but I recall that they are similar to the outback (both controllers designed by the same engineer). Take a look at this thread where I ran the numbers for a system not too different from yours:
    http://forum.solar-electric.com/showthread.php?15907

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • ggunn
    ggunn Solar Expert Posts: 1,973 ✭✭✭
    Re: Another panel series/parallel question
    Chesuncok1 wrote: »
    Thank you for all the input.

    How can you figure the losses of the MidNite Classic 150? I like the idea of 3 panels in a string for 52 volts, how much more would I lose if I went 4 panels in a string for 72 volts?
    You need to figure what your highest Voc (not Vmp) would be for a string of four modules on the record coldest day to make sure you won't exceed the maximum voltage your CC can handle.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Another panel series/parallel question

    Most of the MPPT type controller manufacturers have an online string sizing tool. Here is Midnite's (I have not tried it yet):

    Classic String Sizing Tool


    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Chesuncok1
    Chesuncok1 Registered Users Posts: 5
    Re: Another panel series/parallel question
    vtmaps wrote: »
    I don't have the efficiency curves for the classic, but I recall that they are similar to the outback (both controllers designed by the same engineer). Take a look at this thread where I ran the numbers for a system not too different from yours:
    http://forum.solar-electric.com/showthread.php?15907

    --vtMaps
    Thanks Guys,
    vtmaps, thanks, I checked out the site for Outback and it looks like a difference of 1% between 54 and 72 volts. I will pick up about 1.5% in wire loss and both of these are at full output. So just like that tread discussed, it is fine either way. I will go with 72 volt which will help me when I move everything 30' further away next year.

    ggunn, thanks, my voc with 4 panels in a string is 86.8 volts and I was surprised to see at our cold winter temperature of -30 the voc could be 104 volts! But that is still under the 150 by a good margin.

    BB, thanks, the above numbers that I was telling ggunn about where from the spread sheet at MidNite site that you told me about.

    The 6 new panels arrived today, I will be going to camp to install them in 2 weeks on whenever the MidNite Classic 150 shows up. I build a new pole mount that easily adjusts as the sun angle changes. With all 12 panels (1690 watts) I hope to be good for a while. Next year I hope to purchase a 24 volt 875 amp (6 hr. rating) fork lift battery. Do you guys have any thoughts on fork lift batteries?

    Thank so much, you guys are great.
    Scott
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Another panel series/parallel question
    Chesuncok1 wrote: »
    vtmaps, thanks, I checked out the site for Outback and it looks like a difference of 1% between 54 and 72 volts. I will pick up about 1.5% in wire loss and both of these are at full output. So just like that tread discussed, it is fine either way. I will go with 72 volt which will help me when I move everything 30' further away next year.

    Here's what I see:
    54 volts = four strings of three panels: 57.8 watts dissipated in the 70 ft cable at full power
    or
    72 volts = three strings of four panels: 32.5 watts dissipated in the 70 ft cable at full power

    The 1% difference in efficiency on 1700 watts means that your controller will dissipate an extra 17 watts at 72 volts compared to 54 volts..

    Therefore at full power there is an overall efficiency advantage of 25 - 17 = 8 watts in favor of the higher voltage.

    The question is would you rather dissipate an extra 25 watts in the cable or an extra 17 watts in the controller.

    Remember, these numbers are at full power. The power loss in the cable goes as the square of the power transmitted, therefor at less than full power the slight advantage of the higher voltage will be lost. At anything less than full power you will have greater overall efficiency at the lower voltage AND your controller will run cooler.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Chesuncok1
    Chesuncok1 Registered Users Posts: 5
    Re: Another panel series/parallel question
    vtmaps wrote: »
    Here's what I see:
    54 volts = four strings of three panels: 57.8 watts dissipated in the 70 ft cable at full power
    or
    72 volts = three strings of four panels: 32.5 watts dissipated in the 70 ft cable at full power

    The 1% difference in efficiency on 1700 watts means that your controller will dissipate an extra 17 watts at 72 volts compared to 54 volts..

    Therefore at full power there is an overall efficiency advantage of 25 - 17 = 8 watts in favor of the higher voltage.

    The question is would you rather dissipate an extra 25 watts in the cable or an extra 17 watts in the controller.

    Remember, these numbers are at full power. The power loss in the cable goes as the square of the power transmitted, therefor at less than full power the slight advantage of the higher voltage will be lost. At anything less than full power you will have greater overall efficiency at the lower voltage AND your controller will run cooler.

    --vtMaps

    Thanks vtMaps, sorry it took so long t get back. Sometimes work really gets in the way, had to do a mini marathon.

    Makes complete sense, I will do 4 strings of 3 panels.

    What is your thought on using regular circuit breakers like Square D for DC current?

    I have a combiner box with each string fused, so I am OK there. I only have a automotive type fuse between the controller and the batteries so I would like to replace that with a new circuit breaker box and protect the PV going into the controller and protect the line going to the batteries.

    Thanks again
    Scott
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Another panel series/parallel question
    Chesuncok1 wrote: »
    What is your thought on using regular circuit breakers like Square D for DC current?
    Not OK. Get DC breakers. Look at Midnite, they make a variety of DC boxes and breakers.

    btw, Many DC breakers are polarized. Its not always immediately obvious which way to wire them. for example, the breaker between the solar panels and the controller will have its 'line' (+) side towards the controller and its 'load' (-) side towards the solar panels.
    Makes complete sense, I will do 4 strings of 3 panels.
    It makes sense given these particular numbers. When you extend the distance from 70 ft to 100 ft you may wish to reconsider the configuration, perhaps the higher voltage will make more sense.

    Generally speaking, if you need to make the array voltage more than twice the battery voltage, you are just making the best of a bad situation. The best solution is to either upgrade your 70 ft cable, or switch to a 48 volt system (you could put 6 panels in series).

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Another panel series/parallel question
    vtmaps wrote: »

    btw, Many DC breakers are polarized. Its not always immediately obvious which way to wire them. for example, the breaker between the solar panels and the controller will have its 'line' (+) side towards the controller and its 'load' (-) side towards the solar panels.
    --vtMaps

    Tutorial mode for those who are interested in the why behind the how. The rest of you can just ignore this since vtmaps already gave the correct answer very concisely. :-)

    There are two main reasons that a DC breaker could be considered directional, and they lead to the same definition of which side should be + and which side should be - when the breaker opens during a normal or a fault condition. But the current flow in the two circumstances may well be different.

    1. Since the DC is going to have the electrons (or negative ions during arcing) always going in the same direction during normal conditions, it matters which of the actual contacts is on the + side. They may be made of different materials and/or shaped differently. This affects possible damage to the contacts themselves.
    2. Same thing for the extinguishing of the arc that results during an opening under load, especially during a fault condition. It is extinguishing the arc without waiting for the voltage to go through zero that distinguishes a DC rated breaker from an AC-only breaker. And is the reason that the DC voltage rating of an AC/DC breaker is often lower than the AC rating.

    Now if you consider ONLY the use of the breaker as a switch during normal use, the + side should go toward the panel + and the - side should go toward the charge controller.

    But if you look at a fault condition, in which the breaker will trip because of excess current, that can never happen as a result of the panel output, even if you short across the input of the inverter. The only source of current above the rating of the breaker will be the battery, and that can happen only in the event of a failure in the CC itself that puts battery directly back onto the panels. And it will cause high current only if there is also a short in the panels. You are less likely to have two faults at the same time unless they have a common cause (such as a lightning strike damaging the electronics) but unlikely or not, it is the only fault condition you need to worry about. And in this case the + side of the breaker should be toward the CC (and indirectly the batteries) and the - side of the breaker should be toward the panels.

    Forced to choose between the life of the breaker when used many times to interrupt a small (less than rated) current and the correct operation of the breaker when a really high and dangerous fault current is flowing, we make the second choice.

    Rather than go through this sort of exercise every time you install a DC breaker, you can look to the manufacturer (at least in the case of Midnight Solar) for guidance. And for this situation, they recommend just what vtmaps explained to you.
    SMA SB 3000, old BP panels.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Another panel series/parallel question

    To add--There are other ways of addressing Arcing issues of DC relay/switches--Besides pressurized gas, there is using a permanent magnet to "blow out" the DC Arc (can be polarity sensitive if the arc is "blown" the wrong way).

    Don't know if the breakers have this (or other) polarity sensitivity--But even Load/Source side may be "wrong" if installed between a positive ground (typically telecom systems) vs negative ground (typical consumer) systems.

    Another issue is "remote trip" type breakers... Typically the "power" for remote trip comes from the load side of the breaker.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Chesuncok1
    Chesuncok1 Registered Users Posts: 5
    Re: Another panel series/parallel question
    inetdog wrote: »
    Tutorial mode for those who are interested in the why behind the how. The rest of you can just ignore this since vtmaps already gave the correct answer very concisely. :-)

    There are two main reasons that a DC breaker could be considered directional, and they lead to the same definition of which side should be + and which side should be - when the breaker opens during a normal or a fault condition. But the current flow in the two circumstances may well be different.

    1. Since the DC is going to have the electrons (or negative ions during arcing) always going in the same direction during normal conditions, it matters which of the actual contacts is on the + side. They may be made of different materials and/or shaped differently. This affects possible damage to the contacts themselves.
    2. Same thing for the extinguishing of the arc that results during an opening under load, especially during a fault condition. It is extinguishing the arc without waiting for the voltage to go through zero that distinguishes a DC rated breaker from an AC-only breaker. And is the reason that the DC voltage rating of an AC/DC breaker is often lower than the AC rating.

    Now if you consider ONLY the use of the breaker as a switch during normal use, the + side should go toward the panel + and the - side should go toward the charge controller.

    But if you look at a fault condition, in which the breaker will trip because of excess current, that can never happen as a result of the panel output, even if you short across the input of the inverter. The only source of current above the rating of the breaker will be the battery, and that can happen only in the event of a failure in the CC itself that puts battery directly back onto the panels. And it will cause high current only if there is also a short in the panels. You are less likely to have two faults at the same time unless they have a common cause (such as a lightning strike damaging the electronics) but unlikely or not, it is the only fault condition you need to worry about. And in this case the + side of the breaker should be toward the CC (and indirectly the batteries) and the - side of the breaker should be toward the panels.

    Forced to choose between the life of the breaker when used many times to interrupt a small (less than rated) current and the correct operation of the breaker when a really high and dangerous fault current is flowing, we make the second choice.

    Rather than go through this sort of exercise every time you install a DC breaker, you can look to the manufacturer (at least in the case of Midnight Solar) for guidance. And for this situation, they recommend just what vtmaps explained to you.

    Thanks again to all three of you. I thought there was a difference but my electrician said you can use them on AC or DC. That is why I asked as I have found that most electricians do not have the knowledge of solar systems to give the correct answer. I appreciate the detail explanation as I always like to know why.

    Thanks
    Scott